I've played maybe half a dozen or more PbP games dugin the year, and I have mostly the same feelings as you. All those games lasted at more 2 encounters. The only game that lasted longer than that was the only PbP I DMed myself, which was 2 years to run basically
half a Dungeon Magazine adventure (more or less 5-6 encounters) consisting of a dungeon crawl - the other half of the adventure was supposed to take place in the fortress above the dungeon, but we wrapped the game up after clearing the dungeon only.
1: It may be futile to try to do a huge level 1-20 (ish) adventure path. The game that we are about to complete took us from level 4 to 7 (It's 99.9% sure we will be finished in 1-2 weeks tops). It was a somewhat sizable dungeon, but *clearly* not a full-sized campaign... and yet this took us almost 14 months in real life! So I think that doing one module, or maybe a series of 2-3 small adventures, is very reasonable, but a full campaign (like say one of the big ones published in 5e, or a Paizo adventure path) is probably not achievable. Knowing that there is a clear, achievable conclusion in sight can keep people motivated too.
The absolute biggest issue with PbP IMHO is always wanting to do way too much. The more ambition the DM, the louder the sound the game makes when it crashes.
If you want to do something BIG, remember how long it takes in a
tabletop game (including the time between sessions), and multiply it at least by 10. Then you get an idea about how long it will take in a PbP.
You are absolutely right that having a "clear, achievable conclusion" is the way to go, but it seems that most people want to be reassured that the game can potentially continue forever. I think a
module is a better idea: treat a PbP game as a
project which implies a
conclusion condition. If it
really goes well enough to reach the conclusion, it'll be a piece of cake to pick up some adventure seeds and add another module afterwards.
2: The start of a game is important. There can be a *lot* of energy and excitement at first, and it's important for the GM to take advantage of it and go with it. This energy can be spent designing the characters and the setting, but I think it's important that not TOO much effort be spent on this, because the start of the adventure can then feel like a let down! So starting the actual game with a bit of a bang can be good. I known it's good to have an easy "introduction-battle" to let people gel, but it can be boring. On the other hand, that bang shouldn't be something too railroady. "Oh that idilic town that you guys spent 2-3 weeks thinking about? UNSTOPPABLE INVASION, RUN FOR YOUR LIVES!!!!"... not always the best. (I've had it twice and both games didn't complete. coincidence?)
Once again you're totally right... Every PbP starts with people swearing they'll post every day, writing huge character backgrounds, discussing the details of character generation, talking about house rules, the DM describing her complete fantasy setting... all of which becomes useless as soon as the game swamps at (or before) the first scene or encounter!
I really don't think PbP is the place where you should exercise your character-building or world-building skills. The more effort you put in the design part, the faster your enthusiasm will deplete, and you won't have much left to play the game itself. Keep those for your RL-tabletop gaming groups, and focus instead on getting into the story and action as soon as possible, while your hype for the game is still up.
A few practical suggestions for the DMs:
- use the PbP game to try out a
couple of house rules, but nothing more, otherwise players will be distracted
- forget about detailing your fantasy settings fully in advance, players don't care and won't remember much of it
-
never require players to write long backgrounds or lots of character details: they won't matter if the game doesn't last long, a few sketches are ok, everything else can be added later when (if) the game lasts
Also some for the players:
- stick to the DM's proposals, don't immediately try to deviate from the game premises (e.g. "We play an adventure where you are all X and Y". "Oh, can I play Z? I really really want to play Z!")
- don't bother with optimizing your character, combat will NEVER run in a way that will allow you to take advantage of min-maxing
3: On the topic of railroad, too much choices can paralyze the campaign, so a clear path forward help things move along
Railroading in the open can alienate some players, but I agree that sandboxing is more difficult in PbP. Not because a DM can't improvise in PbP, but because when PbP players don't have a clue, they just don't post anything, and they wait for someone else to make a decision. This can seriously turn the game into a drag.
When you are instead sandboxing in a tabletop game, this is not a problem, simply because the situation in front of the table doesn't allow people to just stare at each other and do nothing for hours, or even minutes... in PbP that stalling could last weeks!
4: Returning to starting a game - I don't think starting at level 1 is great. I think 3 is the minimum, and if I was to run a game tomorrow I would have characters start at level 4 or 5. Fighting kobolds isn't that fun. You don't want to make the players wait to hit that "D&D mid-level sweet spot". If the game won't be super long (see point 1), why not *start* there? Besides, it makes for characters with more robust and varied backgrounds, and multi-class (if you allow it) is more feasible.
That's a good idea, but with a caveat: the higher the starting level, the more time and effort is needed for building the character, and this will work against the principle of trying to get into the game asap.
5: It is quite important that the party gels well. So some kind of "party balance" is nice (doesn't have to be perfect - heck in one game we have 2 fighters, a paladin, a ranger and a monk!), but what I'm talking about is more that the characters get along and aren't too disruptive. This mean that the "fun pirate who's actually kinda evil and a jerk" can be a problem, the "attack everything" barbarian could be a problem, the "ultra loner" is another problem... Heck I have to admit that I was a bit guilty of that in one game - Udit is a pessimist and he was a bit of a pain initially - I realized that was an issue and toned it down (sorry Fitz!). I have seen games where an arrogant, not willing to play along party member sapped the energy of a game and it fizzled out.
I don't think party balance in tactical terms should be a concern. The PbP game will be SHORT, so it's not important to make sure you have a healer, a trapfinder, a tank and so on... just gather up enough PC, and see how they play together tactically. The only reason to require some specific role (such as a trapfinder) is if the adventure expects it in order to surpass specific obstacles.
6: Games with intrigue and politics can be challenging to run. Games where the setting is very different than your "generic late-medieval vaguely European" default can make it more challenging for the players to get engage because there is now a "cultural" barrier - however if you can navigate that, it can be *very* interesting.
Intrigue is
always challenging to run for a DM... we're not Agatha Christie unfortunately!
OTOH if there is one advantage of PbP versus tabletop, is that it gives players a lot more time to think... so personally I think that intrigue/investigation adventures might actually even work better in PbP.
Similarly, another advantage of PbP is that it gives shy players more possibilities for roleplay, that might be taken away by more exuberant players in tabletop. The risk of dominating social encounters exists also in PbP for sure, and quick question-and-answer conversations don't work well, but at least in PbP a player is more free to carefully word her character's speeches if desired.
7: I'm really not sure what is the ideal number of players. Some have suggested over-recruiting to account for player loss/attrition, but at the same time with more people the odds of one player being a bit busy or slow increase, therefore slowing down the game.
Hard to say, but more or less I think it's the same as in tabletop. The smaller number, the easier the management of the group, but the more difficult to beat the adventure
I have yet to understand how a PbP game could be managed really well in terms of pacing everyone's contributions... One thing that could help to avoid the slowness of combat encounters, is to replace cyclic initiative with the old-style declaration of intent per turn, with the DM resolving everything at once.